AirFogComp: Over-the-Air-Fog Computation for Federated Learning over Fog-RAN

This work studies an over-the-air-fog computation (AirFogComp) system, wherein Internet-of- Things (IoT) devices collaboratively learn a machine learning model by communicating with a central server (CS) through a network of access points (APs). Considering the finite capacity of fronthaul links between APs and the CS, we address the challenge of jointly optimizing linear precoding at the IDs, linear processing, and quantization noise covariance matrices at the APs, along with linear combining at the CS. The objective is to minimize the mean squared error (MSE) of the target vector, which is defined as a weighted sum of the local model vectors. To tackle this optimization problem, we propose an iterative block coordinated descent (BCD) algorithm. Numerical experiments demonstrate the rapid convergence of the proposed algorithm and its superior performance compared to baseline schemes.